Visualizations : Go to the red depts

This one may be right.
But the dentists are still not highly paid…maybe due to they have other titles in practice which makes more money:)

Thanks for the update! This view is very interesting, and I wonder how it compares to other schools. The business faculty earns a lot, but the lawyers still beat them by a mile…

The allergists and dentistry folks, who earn very little, are in small departments. Could these be new areas, and the faculty more junior? I wonder if there are stats on average ages…

Hi Martin,
We’d have to find other state schools that are required to post this kind of information (UofM has to, which is why we are using it for the class).
Actually, I have a question about constructing an effective treemap here. Should we have a blank instead of a ‘-’ column for departments that have no subdepartment? I was hoping the students could create something that has internal medicine and pediatrics nested within medicine, but visualizing the subdepartments as the highest level of detail then leaves just ‘-’ for the departments that have none. Thanks! Lada

Hi Lada,

Yep, a blank column is OK. The tricky thing is that it should go in the middle—the labels always come from the last column. This started as a feature, but maybe it’s more like a bug :-) I switched the columns around and uploaded a new data set, and you can see the result here by clickingon this image… let me know if I’ve misinterpreted your question, though!

Thanks Martin, that is just the thing that it needed. Another treemap visualization I’ve used before (http://www.cs.umd.edu/hcil/treemap/) has an optional number of nested categories at the end of each line, e.g.
school1
school2 dept2 subdept2
school3 dept3
…
and I think this allows it to figure out what label to use.

This led me astray in trying to use subcategories here as well. In any case, thanks much for helping us out! This is exactly the visualization we were after.
(I also tried uploading a dataset that set the subdepartment column to the department column for those entries that had no subdepartments, but this resulted in duplicate cells…)